294
Chem. Res. Toxicol. 2000, 13, 294-300
4-Hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl (Tempol) Inhibits Peroxynitrite-Mediated Phenol Nitration Richard T. Carroll,*,† Paul Galatsis,‡ Susan Borosky,† Karla K. Kopec,† Vikram Kumar,† John S. Althaus,† and Edward D. Hall† Neuroscience Therapeutics, Parke-Davis Pharmaceutical Research, Division of Warner-Lambert Company, 2800 Plymouth Road, Ann Arbor, Michigan 48105, and Neuroscience Chemistry, Parke-Davis Pharmaceutical Research, Division of Warner-Lambert Company, 2800 Plymouth Road, Ann Arbor, Michigan 48105 Received September 3, 1999
Peroxynitrite (PN), a very reactive oxidant formed by the combination of superoxide and nitric oxide, appears to play a role in producing tissue damage in a number of inflammatory conditions. Pharmacological scavenging and decomposition of PN within these areas has therapeutic value in several tissue injury models. Recently, we have been interested in nitroxide free radical-containing compounds as possible scavengers of PN decomposition products. Nitroxides can undergo redox reactions to the corresponding hydroxylamine anion or oxoammonium cation in biological systems as shown by its ability to react with superoxide, leading to the formation of hydrogen peroxide and molecular oxygen. We found that 4-hydroxy-2,2,6,6tetramethylpiperidine-1-oxyl (Tempol) inhibits PN-mediated nitration of phenolic compounds in the presence of a large molar excess of PN, suggesting a catalytic-like mechanism. In these experiments, Tempol inhibited PN-mediated nitration over the pH range of 6.5-8.5. This inhibition was specific for nitration and had no effect on hydroxylation. After the inhibition of PN-mediated nitration, Tempol was recovered from the reaction mixtures unmodified. In addition, Tempol was effective in protecting PC-12 cells from death induced by SIN-1, a PNgenerating compound. The exact mechanism of Tempol’s interaction with PN is not clear; however, we propose that an intermediate in this reaction may be a nitrogen dioxide radicalTempol complex. This complex could react with water to form either nitrite or nitrate, or with a phenol radical to produce nitrophenol or nitrosophenol products and regenerate the nitroxide.
Introduction Peroxynitrite (PN) is a very reactive oxidant formed by the combination of superoxide with nitric oxide (1). The reaction between superoxide and nitric oxide occurs at the limit of diffusion control with a second-order rate constant of 1.9 × 1010 M-1 s-1 (2), which is faster than the enzymatic reaction of superoxide dismutase with superoxide. Compared to other biological reactive oxygen species, PN is relatively long-lived with a t1/2 of about 1 s at physiological pH and is consequently capable of crossing cellular membranes and attacking extracellular as well as intracellular targets (3, 4). PN can lead to tissue damage via the nitration of tyrosine residues (5), oxidation of methionine (6) and cysteine residues (7), and DNA strand breakage (8-10). The consequence of these reactions can compromise cellular homeostasis and lead to cell death. The production of PN has been shown to be associated with multiple neurological diseases and may play a role in the pathology of disease progression. In brains from Alzheimer’s disease patients, nitrotyrosine and other * To whom correspondence should be addressed: Neuroscience Therapeutics, Parke-Davis Pharmaceutical Research, 2800 Plymouth Rd., Ann Arbor, MI 48105. Phone: (734) 622-5216. Fax: (734) 6227178. E-mail:
[email protected]. † Neuroscience Therapeutics, Parke-Davis Pharmaceutical Research, Division of Warner-Lambert Co. ‡ Neuroscience Chemistry, Parke-Davis Pharmaceutical Research, Division of Warner-Lambert Co.
markers of oxidative damage are associated with areas of neuronal degeneration (11-16). Nitrotyrosine accumulation is also evident in brains obtained from patients suffering from multiple sclerosis (MS) (17, 18). It is thought that PN production from infiltrating activated macrophages may play a part in the formation of brain lesions in MS patients. In fact, a scavenger of PN, uric acid, prevents the development of experimental allergic encephalomyelitis in mice, suggesting possible therapeutic applications in human MS (19, 20). PN is also strongly implicated in head trauma (21), brain ischemia (22-24), and amyotrophic lateral sclerosis (25). In addition, PN may amplify the immune response in these diseases through the activation of prostaglandin biosynthesis (26), causing increased collateral damage to surrounding tissues. Under physiological conditions, PN can react with carbon dioxide to form nitrosoperoxocarbonate anion (ONOOCO2-) at a rate constant of 5.8 × 104 M-1 s-1 (27). Considering the high concentration of carbon dioxide (1-2 mM) in the extracellular and intracellular spaces (28), it is likely that the majority of PN formed in vivo reacts with CO2 to form ONOOCO2- (29). The CO2 adduct of PN has altered reactivity, giving rise to an increased level of nitration of aromatic residues and slower reactivity with antioxidants such as glutathione (27, 30, 31). In the case of phenol nitration by PN, a 2-fold increase in the level of nitrated products was obtained when bicar-
10.1021/tx990159t CCC: $19.00 © 2000 American Chemical Society Published on Web 03/15/2000
Tempol Inhibits Phenol Nitration
bonate was added to the reaction mixture (29). These observations suggest that scavenging the CO2 adduct of PN would be useful in reducing tissue damage in areas of high nitric oxide production. The focus of this work was to characterize a new type of PN scavenger. The nitroxide, 4-hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl (Tempol), was found to act as an inhibitor PN-mediated nitration and an inhibitor of cell death induced by the PN generator SIN-1.
Materials and Methods Synthesis of Peroxynitrite. The synthesis of PN was performed as previously described with slight modifications (32). Briefly, a pH 12 solution of 0.2 M NaN3 (100 mL) in a 100 mL glass bubbler was ozonized at 0 °C using a Welsbach ozonator (model T-23) operating with an oxygen pressure of 8 psi. A flow rate of 0.05 SCFM and an electric discharge of 90 V were employed. This typically gave 80 mM PN solutions in about 2530 min. The concentration of PN was determined spectrophotometrically by diluting the sample 100-fold in water (pH 12) and measuring its absorbance at 302 nm (302 ) 1670 M-1 cm-1). Nitration of 4-Hydroxyphenylacetate by Peroxynitrite. Solutions of 4-hydroxyphenylacetate (HPA, 1 mM, Aldrich) were made in 100 mM sodium phosphate at the indicated pH and sodium bicarbonate concentrations. HPA solutions (100 µL) were mixed with an equal volume of PN (2.5 mM in H2O) to give a final PN concentration of 1.25 mM. Reactions were performed in a 96-well plate, and nitration was followed spectrophotometrically at 405 nm using a microplate reader (Molecular Devices, Thermomax). HPA nitration inhibition experiments were performed by mixing Tempol, at the indicated concentrations, in the HPA solutions prior to PN addition. The concentration of 4-hydroxy-3-nitrophenylacetate was determined spectrophotometrically (430 ) 4400 M-1 cm-1) after the pH of the reaction mixtures had been increased to 10-11 with 1 N NaOH (33). HPLC Analysis of Phenol and Its PN-Mediated Nitration Products. Solutions of phenol (1 mM, Aldrich) were made in 100 mM sodium phosphate at the indicated pH containing 50 mM sodium bicarbonate with and without 25 µM Tempol. Phenol solutions (100 µL) were mixed with an equal volume of PN (2.5 mM in H2O) to give a final PN concentration of 1.25 mM. Reactions were performed in a 96-well plate, and nitration was followed spectrophotometrically at 405 nm using a microplate reader (n ) 6, Molecular Devices, Thermomax). Product analysis was performed using a modified procedure described by Lemercier et al. (34). Briefly, reaction products were analyzed using a Rainin HPXL gradient system (Rainin Instrument Co., Inc.) with a dual-wavelength SpectroMonitor 4100 detector (Thermoseparations Products). Separations were carried out on a Nucleosil 5 C18 100A column (150 mm × 4.6 mm, Phenomenex) using mobile phases A [27 mM acetate/30 mM citrate buffer (pH 3.2)] and B (acetonitrile). A gradient of 15 to 52% B was employed over the course of 18 min at a flow rate of 1 mL/ min. The column was then washed for 2 min with 100% acetonitrile and equilibrated at 15% B for 15 min. The peak area (milliabsorbance units per second) was measured from chromatographs obtained from absorbance measurements at 280 nm, and retention times were compared to standards for peak identification. Effects of Tempol on PN-Induced Phenol Hydroxylation. Solutions of phenol (10 mM, Aldrich) were made in 100 mM sodium phosphate (pH 6.5) with and without 25 µM Tempol. Phenol solutions (100 µL) were mixed with an equal volume of PN (1.0 mM in H2O) to give a final PN and phenol concentrations of 500 µM and 5 mM, respectively. Reactions were performed in a 96-well plate, and nitration was followed spectrophotometrically at 405 nm using a microplate reader (n ) 6, Molecular Devices, Thermomax). Reaction products were analyzed by HPLC as described above using the modified procedure described by Lemercier et al. (34).
Chem. Res. Toxicol., Vol. 13, No. 4, 2000 295 Tempol Recovery after Its Inhibition of Phenol Nitration. Solutions of phenol (1 mM, Aldrich) were made in 100 mM sodium phosphate (pH 7.5) containing 50 mM sodium bicarbonate and either 0, 6.25, or 12.5 µM Tempol. Phenol solutions (100 µL) were mixed with an equal volume of PN solution (1 mM in H2O) to give a final PN concentration of 0.5 mM. Controls were run by mixing phenol solutions with an equal volume of H2O. Reactions were performed in a 96-well plate, and nitration was followed spectrophotometrically at 405 nm using a microplate reader (n ) 6, Molecular Devices, Thermomax). The mixtures described above were examined for Tempol content using LC/MS/MS. The system consisted of a Waters 2790 Separations Module for compound separation prior to analysis by a Quattro Ultima tandem mass spectrometer (Micromass). Separation was carried out using a TSK-GEL column (TosoHaas, ODS-80TS, 2 mm × 15 cm, 5 µm particle size). Samples were maintained at 4 °C prior to injection. Chromatography was carried out isocratically at 30 °C using a mobile phase consisting of 15% acetonitrile in 0.1% acetic acid and a flow rate of 0.270 mL/min. The eluent from the LC system was then analyzed using an electrospray ionization (ESI) technique. The mode examined was positive ionization (ESP+) ion selection and multiple reaction monitoring (MRM). The parent ion (M + 1) had a mass of m/z 173.10; the daughter ion that was selected had a mass of m/z 158.04. The dwell time for the scan was 0.25 s. The cone voltage was 28 V, and the collision energy was 14 eV. The source block temperature was 150 °C, and the desolvation temperature was 270 °C. The cone gas rate was set at 204 L/h, while the desolvation gas rate was set at 588 L/h. The LC/MS/MS run time was 10 min. The retention time of Tempol was 6.95 ( 0.5 min. Raw data were processed using the MassLynx software package. A calibration curve was created by running Tempol standards at 1, 0.1, 0.01, and 0 µg/mL. The areas of the analyte peaks were then measured, and their concentrations in micrograms per milliliter were found using the calibration curve as a reference. Tempol-Mediated Nitrite Formation from Peroxynitrite. Solutions of 2 mM HPA were made in 100 mM sodium phosphate (pH 7.5) containing 50 mM sodium bicarbonate containing the indicated amount of Tempol. HPA solutions (100 µL) were mixed with an equal volume of 1.0 mM PN to give a final PN concentration of 0.5 mM. Control experiments were performed using the buffer described above with the indicated Tempol concentration and no HPA. The control samples were also mixed with an equal volume of 1 mM PN. Reactions were performed in a 96-well plate and the mixtures allowed to set for 5 min at room temperature. The samples were then diluted 1:10 in H2O. The diluted samples were then assayed for nitrite concentration using a modified method of the Griess assay (35). Briefly, 100 µL of diluted sample was mixed with 50 µL of 1% sulfanilamide (Sigma) in 0.1 M HCl and 50 µL of 0.1% N-(1naphthyl)ethylenediamine (Sigma) in H2O. Samples were allowed to set at room temperature for 5 min, and the absorbances were read at 550 nm. Sample concentrations were extrapolated from a standard curve made from nitrite standards ranging from 0 to 250 µM. Contaminating nitrite in the PN solutions was evaluated by mixing 1 mM PN with an equal volume of 5% phosphoric acid prior to nitrite analysis. The background level of contaminating nitrite was then subtracted from all experiments. Cell Culture and Peroxynitrite-Induced Cytotoxicity. Cytotoxicity was induced in PC12 cells following the procedures of Estevez et al. (36) with minor modifications. In collagencoated 96-well plates, PC12 cells were plated at a density of 1.5 × 105 cells/cm2 in OptiMem (Gibco BRL) containing 4% fetal bovine serum and 5 µg/mL gentamicin (Gibco BRL) and allowed to adhere overnight. Cultures were then washed once with phosphate-buffered saline containing 5 mM glucose (pH 7.4) and then placed in 200 µL of OptiMem containing 5 µg/mL gentamicin. 3-Morpholinosydnonimine (SIN-1, Cayman Chemicals), a PN generator, was dissolved in 50 mM phosphate (pH 5.0) just
296
Chem. Res. Toxicol., Vol. 13, No. 4, 2000
Carroll et al. Table 1. HPLC Peak Areas of Products Obtained from PN Nitration of Phenola
Figure 1. Inhibition of PN-mediated nitration of 4-hydroxyphenylacetate by Tempol. The reactions were carried out using 0.5 mM HPA, 1.25 mM PN, and 25 mM bicarbonate in 50 mM sodium phosphate. Each point represents six iterations plotted with the standard deviation. Symbols represent the pH at which the reactions were carried out: (O) pH 6.5, (0) 7.0, (]) 7.5, (9) 8.0, and ([) 8.5.
pH
4-nitrosophenol
phenol
4-NP
2-NP
[Tempol] (µM)
8.5 8.0 7.5 7.0 6.5 8.5 8.0 7.5 7.0 6.5
215 46 13 N/Db N/D 396 284 171 139 133
861 826 729 609 580 955 902 942 975 968
58 121 526 765 846 N/D N/D N/D N/D 41
128 265 1010 1130 1148 43 N/D N/D N/D 36
0 0 0 0 0 12.5 12.5 12.5 12.5 12.5
a Reaction conditions are identical to those reported in the legend of Figure 2. Peak areas (millivolts per second) were obtained from HPLC chromatograms while monitoring absorbance at 280 nm. b N/D, not determined. Peaks fell below our ability of detection for accurate quantitation.
prior to use, and 5 µL was added to each well to give the indicated concentration. The cells were allowed to incubate for 24 h, and cytotoxicity was determined using a detection kit (Boehringer Mannheim) that measured lactate dehydrogenase (LDH) activity released from damaged cells. In experiments with the PN scavenger, Tempol was added at the indicated concentrations to each culture 15 min prior to SIN-1 (1 mM final concentration) addition. SIN-1 at a concentration of 1 mM has been reported to generate nitric oxide and superoxide at rates of 3.68 and 7.02 µmol/min, respectively (37). The percent inhibition of PN-mediated cytotoxicity was determined by the percent reduction in LDH activity release as compared to that of SIN-1-treated cultures. As a control, 50 units/mL SOD (Sigma, S-2515) was added to SIN-1-treated cultures to prevent the formation of PN.
Results Inhibition of 4-Hydroxyphenylacetate Nitration by Tempol. The ability of Tempol to act as a scavenger for PN-mediated nitration was examined using the HPA nitration assay. Tempol protected 500 µM HPA from nitration by 1.25 mM PN in a concentration-dependent manner (Figure 1). Under these conditions, Tempol was able to prevent PN-mediated nitration with IC50 values of 2.5, 3.2, 3.7, 5.5, and 8.4 µM at pH 6.5, 7.0, 7.5, 8.0, and 8.5, respectively. Maximal nitration occurred at pH 6.5 and resulted in the nitration of approximately 105 µM HPA. The 20-fold turnover of Tempol under these conditions suggests that it is acting in a catalytic-like fashion. Effect of Alkaline pH on Tempol’s Interactions with Phenol and Peroxynitrite. Phenol nitration and nitrosation by PN were evaluated by HPLC at pH 6.5, 7.0, 7.5, 8.0, and 8.5 (Table 1). The maximum amount of phenol nitration in this experiment occurred at pH 6.5 and decreased with increasing pH. Conversely, increased nitrosation of phenol occurred as the pH increased. In our experiments, 4-nitrosophenol represented approximately 17, 4, and 1% of the total peak areas at pH 8.5, 8.0, and 7.5, respectively (Figure 2A). When Tempol was added to the reaction mixtures, the nitration of phenol was almost completely abolished under all conditions (Figure 2B). However, Tempol caused increases in the extent of formation of nitrosophenol of 28, 24, 15, 13, and 11% at pH 8.5, 8.0, 7.5, 7.0, and 6.5, respectively (Table
Figure 2. HPLC profiles of products obtained from PN reactions with phenol. Reactions were carried out using equal volumes of 1.0 mM phenol in 50 mM sodium phosphate and 50 mM sodium bicarbonate at the indicated pH with 2.5 mM PN. The reaction mixtures contained no Tempol (A) or 12.5 µM Tempol (B). Peak identities were as follows: peak 1, 4-nitrosophenol; peak 2, phenol; peak 3, 4-nitrophenol; and peak 4, 2-nitrophenol. Product identification was made by comparison of HPLC standards separated under identical conditions.
1). An increase in the amount of nitrosated product implies the formation of a complex between Tempol and one or more of the decomposition products of PN. Effect of Tempol on PN-Mediated Hydroxylation of Phenol. When using a 10-fold molar excess of phenol over PN at pH 6.5, hydroxylated products of phenol, as well as some minor nitration products, were identified by HPLC (Figure 3). However, these hydroxylated products were not formed in the presence of added bicarbonate (Table 2). When 12.5 µM Tempol was added to the reaction mixture, the formation of nitrated products (2and 4-nitrophenol) was inhibited and was independent of whether bicarbonate was added to the reaction mixture. Conversely, the extent of formation of hydroxyphenol products slightly increased in the presence of Tempol (Figure 3 and Table 2). Tempol also increased the extent of formation of nitrosophenol with or without the addition of bicarbonate to the reaction mixture (Table 2).
Tempol Inhibits Phenol Nitration
Chem. Res. Toxicol., Vol. 13, No. 4, 2000 297
Figure 3. HPLC profiles of PN-mediated hydroxylation of phenol. Solutions of 100 mM sodium phosphate (pH 6.5) containing 10 mM phenol were made without (A) and with (B) 25 µM Tempol. Phenol solutions were mixed 1:1 with 1.0 mM PN 10 min prior to HPLC analysis. Product identification was made by comparison of HPLC standards separated under identical conditions (C). Peak identities were as follows: peak 1, hydroquinone; peak 2, catechol; peak 3, 4-nitrosophenol; peak 4, phenol; peak 5, 4-nitrophenol; and peak 6, 2-nitrophenol.
Figure 4. Tempol recovery after its inhibition of phenol nitration. Reactions were carried out using equal volumes of 1.0 mM phenol in 100 mM sodium phosphate and 50 mM sodium bicarbonate (pH 7.5) containing either 6.25 or 12.5 µM Tempol with 1 mM PN. The percent recovery of Tempol, as compared to those of the non-peroxynitrite-treated samples, was determined by HPLC/MS/MS (solid bars). The extent of inhibition of phenol nitration in these reactions was calculated by the decrease in absorbance at 405 nm as compared to those of nonTempol-treated reactions (hatched bars). Each sample represents an average of six iterations with the standard deviation.
Table 2. HPLC Peak Areas of Products Obtained from PN Hydroxylation of Phenola HCO3 0 0 25 25
hydroxy4-nitroso[Tempol] quinone catechol phenol 4-NP 2-NP (µM) 40 46 N/Db N/D
62 89 N/D N/D
N/D 22 N/D 55
24 N/D 153 34
44 N/D 292 65
0 12.5 0 12.5
a Reaction conditions are identical to those reported in the legend of Figure 3. Peak areas (millivolts per second) were obtained from HPLC chromatograms while monitoring absorbance at 280 nm. b N/D, not determined. Peaks fell below our ability of detection for accurate quantitation.
Tempol Recovery after Inhibition of PN-Mediated Phenol Nitration. To help determine the mechanism by which Tempol prevents phenol nitration, Tempol was analyzed for potential modification after inhibiting PN-mediated nitration of phenol. Tempol, at 6.25 and 12.5 µM, inhibited nitration by 67 and 72%, respectively (Figure 4). In this reaction, PN was in a 40- and 80-fold molar excess over Tempol. After inhibition, Tempol was recovered by LC/MS/MS and compared to controls that had no PN added (Figure 4). Peroxynitrite treatment had no significant effect on the recovery of Tempol, demonstrating that it is not being consumed during its inhibition of PN-mediated nitration. Tempol-Mediated Nitrite Formation from Peroxynitrite. Upon the decomposition of PN, nitrate and nitrite are formed under physiological conditions. The influence of Tempol on this decomposition was evaluated by measuring the extent of nitrite formation in the presence and absence of a HPA. Under these reaction conditions, a solution containing 1 mM HPA and 500 µM PN resulted in the formation of 186 µM nitrite. When 12.5 µM Tempol was added prior to PN addition, the concentration of nitrite that was detected increased to 279 µM. This increase of approximately 93 µM nitrite primarily occurred at Tempol concentrations of
